The objectives of this grant, which has been funded for almost 35 years, have been to understand the mechanisms for the establishment and alteration of the human epigenome in cancer by focusing on DNA methylation patterns. We also strive to understand how drugs such as 5-aza-2'-deoxycytidine (5-aza-CdR) interfere with cytosine methylation and reactivate silenced genes. This research has led to the approval by the FDA of two DNA demethylating agents (5-aza-CR and 5-aza-CdR) for the treatment of myelodysplastic syndrome (MDS) and we have been successful over the last five years in the preclinical development of the first next-generation nucleoside inhibitor of DNA methylation called SGI110. This drug has been undergoing Phase I/II testing and shows great promise. In the next five year period of the project, we hope to take advantage of an innovative new technique we have developed called NOMe-seq which allows us to simultaneously map nucleosomes and DNA methylation patterns either at the level of individual DNA molecules (i.e., replicas of individual transcription start sites, etc.), or genome-wide. By coupling this technique with ChIP-seq and RNA expression analysis and next-generation sequencing, we gain a holistic view of the human epigenome and how it is altered in cancer for the first time. In Specific Aim 1, we will use this approach to define how DNA methylation acts genome-wide to position nucleosomes at regulatory regions and throughout the genome. We will then explore the mechanisms by which chromatin remodelers and histone methyltransferases alter the epigenome. This is crucial because a large number of mutations in chromatin modifiers have been discovered over the past five years, but we know little about the potential effects of these mutations on the epigenome. In Specific Aim 2, we investigate how perturbing DNA methylation with 5-aza- CdR alters the entire epigenome and the expression of subsets of genes. We are particularly interested in exploring the time course by which different sets of genes become activated and then become resilenced as a function of time after treatment. A novel part of this Aim will be to investigate the effect of gene body methylation on transcriptional elongation which may, unexpectedly, play a role in the patient's response to these drugs. In Specific Aim 3, we plan a first epigenome-wide analysis of uncultured human tumors by focusing on colon tumors and adjacent tissue. The completion of these Specific Aims will give us a better understanding of why and how the epigenome is altered in cancer, how it responds to drugs and hopefully will result in better treatments for patients.